deepdrivemd.data.stream.OutlierDB
Classes
|
Stores the metadata for outliers to be used by simulations. |
- class deepdrivemd.data.stream.OutlierDB.OutlierDB(dir: str, restarts: List[Tuple[float, str]])
Stores the metadata for outliers to be used by simulations.
- dir
directory with published outliers
- Type
str
- sorted_index
list of md5sums of outlier positions (used as a name of an outlier pdb or numpy file) sorted by the corresponding rmsd
- Type
List[str]
- dictionary
maps md5sum to rmsd
- Type
Dict
- __init__(dir: str, restarts: List[Tuple[float, str]])
Constructor
- Parameters
dir (str) – directory with published outliers
restarts (List[Tuple[float, str]]) – list of outliers given as tuples of rmsd and md5sum of positions (used as a file name)
- next_random(m: Optional[int] = None, alpha: int = 1, beta: int = 25) str
Return next outlier using beta distribution that prefers smaller rmsds
- Parameters
m (int, default = None) – if m is not None, restrict the random selection to the first m elements of softed_index, otherwise - any element can be chosen.
alpha (int, default = 1)
beta (int, default = 25) – alpha and beta are parameters of beta distribution.
- print(n: int = 5)